Search Results

This chapter establishes the empirical and theoretical goals of the book. It introduces the theoretical backdrop of Anti-contiguity Theory, situating the proposal within a framework inspired by Richards’s (2010, 2016) Contiguity Theory. The chapter previews the book’s main findings and conclusions, while providing details about the book’s chapter-by-chapter organization. It alludes to a number of other issues and questions raised by the proposal and delineates which ones will be dealt with in the book and which will not. The chapter also introduces the book’s running theme—sometimes “syntax” is syntax, and sometimes “syntax” is phonology.

This chapter focuses on the syntax and prosody of wh- in-situ in Krachi. In Krachi, wh- in-situ is available in both root and embedded contexts. Prosodically, embedded complement clauses in the language are parsed as independent Intonational Phrases. The distribution of wh- in-situ in the language is considered against the backdrop of Richards’s (2010, 2016) Contiguity Theory. Given the prosodic facts previously mentioned, Contiguity Theory is shown to be incapable of accounting for long-distance wh- in-situ in the language, motivating an “anti-contiguity” theory of the prosody of wh- and C at the syntax-phonology interface.

This chapter develops the Anti-contiguity proposal based on the distribution of wh- in-situ in Krachi, Bono, Wasa, and Asante Twi. Embedded wh- in-situ is only allowed in Krachi and Bono. The chapter argues that a prohibition on wh- items phrasing with overt C at the level of Intonational Phrase (iP) underpins this distributional asymmetry. In these languages, the acceptability of embedded wh- in-situ correlates with the prosodic status of the immediately containing clause—embedded clauses are independent iPs in Krachi and Bono, but not in Wasa and Asante Twi. Thus, iP boundaries divide C from embedded interrogatives in Krachi and Bono, preventing the items from forming iP constituents. No such boundaries intervene between embedded C and wh- in Wasa and Asante Twi, yielding prosodic mappings in which the items phrase together. Consequently, embedded wh- in-situ is prosodically licit in Krachi and Bono, but not in Wasa or Asante Twi.

This chapter furnishes additional support for the Anti-contiguity theory of wh- prosody by demonstrating that it derives two surprising and mysterious asymmetries of Nupe wh- syntax. The first asymmetry concerns the fact that in embedded clauses the structurally lowest wh- item in a multiple wh- question may not appear in-situ. The second asymmetry concerns the availability of embedded non-interrogative focus and the impossibility of embedded interrogative focus. Both of these asymmetries are argued to be explainable in terms of the Anti-contiguity ban on contiguous wh- phrasing with overt C at the Intonational Phrase level, given that overt embedded C does not introduce an Intonational Phrase boundary in Nupe, as in Wasa and Asante Twi. As a consequence, no Intonational Phrase boundary insulates focused embedded wh- items from overt embedding complementizers, therefore running afoul of the Anti-contiguity prohibition.

This chapter concludes the book by considering Anti-contiguity in a cross-linguistic context. It is shown that the proposal can be successfully applied to derive asymmetries in wh- in-situ distribution beyond the West African languages considered in Chapters 2–4. The chapter focuses on thirteen languages from diverse language families (Romance, Bantu, and Indo-Aryan, among others) and considers the implications of data from these languages for the final formulation of the Anti-contiguity condition. On the basis of these considerations, the Anti-contiguity constraint is parameterized. Among the languages considered against the backdrop of the Anti-contiguity proposal in this chapter are French; Spanish; Catalan; Zulu; Bàsàá; Duala; Shona; Lubukusu; Kiitharaka; Hindi-Urdu; Bangla; Iraqi Arabic; and Malayalam.